asrin

samimi oluyor derken mesafe koyabilir de

[asrin @asriintncy]

Jorge Jesus:
“Şampiyonlar Ligi’nde iki Türk takımı var. Fenerbahçe, Avrupa’da daha kaliteli bir takım bunu dün gördük ve son 15 yılda ülke puanına en fazla katkı yapan Türk kulübü. #BeşiktaşınMaçıVar #isteğebağlı2ilDışıBakanTekin $CHUMP

[Hikmettuncay @karakalemhikmet]

Owning an NFT is one thing.
Being able to actually use it is another.

That’s something I’ve been thinking about for a while in the NFT space.

Being rare doesn’t automatically make something useful.

That’s why @hoodminers_rh caught my attention.

The 5,000 Miners aren’t just sitting there as collectibles. Each Miner has its own smart wallet, creating a direct connection between the NFT and the assets it can interact with.

Free mint.
Activate your Miner.
Then participate in the ecosystem and access its reward mechanics.

That’s the part I find interesting.

Instead of only asking “How rare is it?”, we start asking:

“What can this NFT actually do for its owner?”

Having assets like Bitcoin potentially involved on the rewards side makes the model even more interesting.

Maybe the next step for NFTs isn’t better-looking JPEGs.

Maybe it’s more useful NFTs.

Can HoodMiners actually prove that model works?

@NucleusCodes

[asrin @asriintncy]

Fenerbahçe şanpiyonlar liginde 3 puan kaçırdı
İsmail Kartal istifa etti Aziz Yıldırım kabul etmedi Guardiola teklif götürülür mü
#kalpkrizi #kpsshaksızlığı #FenerinMaçıVar #YanındayızİsmailKartal

[Hikmettuncay @karakalemhikmet]

Good morning legends ☀️

I’ve been thinking about something this morning.

A normal NFT is often like buying an empty house.

Nice to look at, maybe rare, but at the end of the day it’s still just sitting there.

@hoodminers_rh takes a different approach.

Think of each Miner like a small digital mining operation.

It has its own wallet, rewards can accumulate inside it, and when the Miner changes hands, the wallet and what it has earned move with it.

That changes the way I look at the NFT.

The image is just the identity.

The real story is what the Miner manages to build over time.

5,000 free mints
9 rarity levels
Robinhood Chain

That’s the part I’m watching.

Because if an NFT can carry its own history and accumulated value, maybe we’re finally moving beyond NFTs that simply sit in a wallet.

Let’s see what these Miners can dig up ⛏️

@NucleusCodes

[asrin @asriintncy]

bugün galatasaray
#SCPvGS Yunus Akgün Sporting - Galatasaray Mansur Yavaş #kpsshaksızlığı #ChampionsLeague

[Hikmettuncay @karakalemhikmet]

Good morning pookies ☀️

The time to watch is over
Now it’s time to position 🫡

@hoodminers_rh is coming to OpenSea on September 15 🌊

And there are 1,031 GTD WL spots up for grabs through the @NucleusCodes × @hoodminers_rh leaderboard 👀

Contribution matters
Reputation matters
Your position matters

I’m already climbing the leaderboard 🤠

If you’re still watching from the sidelines

You’re already late

Let’s climb 🚀

[Hikmettuncay @karakalemhikmet]

1 gün kaldı
Ve hazırım

@world_xyz'yi çok önceden takip ettim. Bu, 6 haneli airdrop alacağım anlamına mı geliyor?

PS Hâlâ bu seksi rozeti bekliyorum

[Hikmettuncay @karakalemhikmet]

$150,000 paid to creators in one month. 👀

That number alone caught my attention.

But what really interests me about @creator_wire is where that money comes from.

Brands are running real campaigns.
Creators are actually doing the work.
Campaigns get completed.
And creators get paid on time.
That’s the kind of creator economy I want to be part of.
Not just chasing campaigns, but building, creating and getting rewarded for real work.
I’m already here creating.

Now I’m curious how far Creator Wire can take this

[asrin @asriintncy]

yüz kere izlediğimiz ilk bölümün özetinde bile bizim keyif #tuzlukahve #YandaşTffTaraflıSevk #KpssHaksızlıgı 9 Eylül #AdidasıBoykotEt

[asrin @asriintncy]

yüz kere izlediğimiz ilk bölümün özetinde bile bizim keyif #tuzlukahve #YandaşTffTaraflıSevk #KpssHaksızlıgı 9 Eylül #AdidasıBoykotEt

[Hikmettuncay @karakalemhikmet]

The longer I spend around @NucleusCodes the more I realize something:

The real product isn’t content. It’s contribution.

Web3 is already flooded with endless noise. Everyone is posting, everyone is farming attention. But who is actually moving the needle?

That’s why systems that reward real, measurable value instead of pure hype are a game changer.

Perfect example: The new Contribution + Reputation Leaderboard for @hoodminers_rh drops today at 12 PM UTC!

With a massive 1,031 GTD WL spots up for grabs across both leaderboards. Attention can be faked with a template, but real contribution gets rewarded.

Check it out and lock in your spot here:
👉 https://t.co/xvm6hx56rk...

[asrin @asriintncy]

@jonaskoyas Active

[asrin @asriintncy]

@Mi_Zu_Ki_20 Yes

[asrin @asriintncy]

7 di yaşında sınavlara başadık yaş 25 oldu hayat sınavı bitmediği gibi iş girmek atanmak içinde sınava giriyoruz
#kpsslisans #EKPSS #yangın Bülent Korkmaz

[Hikmettuncay @karakalemhikmet]

Today was honestly a big disappointment for me in Web3.

I spent almost 3 months working on @CNPYNetwork for the airdrop.
I completed tasks, created content, and even minted my own token.
But today I found out that my wallet wasn’t eligible.

Of course it hurts when you put that much time and effort into something and get nothing back.

Still, I’m not going to stop building.

Not every effort gets rewarded.
Not every airdrop goes your way.

But I’ll keep creating, keep learning and keep showing up.
The airdrop might be gone, but the journey isn’t.

Still here. Still building. 🫡

@NucleusCodes

[asrin @asriintncy]

@fuyu_kasumii yes

[Hikmettuncay @karakalemhikmet]

Sanki Pandora’nın kutusunu açmışım gibi geliyor!! 👀

@termix_ai ilk günden beri takip ediyorum.

Başta sadece AI agent’ları anlamaya çalışıyordum.

Ama kutuyu biraz daha açtıkça karşıma başka bir şey çıkıyor:

Agent economy.

Ve Eylül ayında bunun birkaç önemli işaretini gördük 👇

🟢 https://t.co/toJ5CI1DbM artık Robinhood Chain’de

Bir agent artık sadece mint edilen bir profil değil.

İş ilanı yayınlayabiliyor
İşe alınabiliyor
Çalışabiliyor
Ve yaptığı iş karşılığında onchain ödeme alabiliyor.

BNB Chain
Base
Robinhood Chain

Agent’ların farklı zincirlerde gerçek ekonomik faaliyetlere katılabildiği bir yapı oluşuyor.

📣 Kaito Katalyst tarafı da büyüyor

700+ creator
Milyonlarca impression
Epoch 1 tamamlandı
Epoch 2 devam ediyor.

Ben de ilk günden beri kendi dilimde TermiX’i anlatıyorum.

Ama burada benim için en değerli şey sadece sıralama değil.

TermiX sayesinde birçok insanla tanıştım.

Bazı insanların AI agent’larını projenin kendisinden bile daha iyi anlattığını gördüm.

İşte o zaman şunu düşündüm:

Belki de Pandora’nın kutusundan çıkan asıl şey agent’lar değil
bu ekonomiyi birlikte kuran insanlar.

Ben hikâyenin nereye gideceğini görmek için buradayım.

Senin de ilgini çekiyorsa
gel birlikte inşa edelim.

Kaito tarafında da beraber ilerleyebiliriz 👇
https://t.co/Wq1BbmjOda

Belki bir sonraki seçilen yorum seninki olur 👀

⏳ Son düzlük: $10K BNB Chain × TermiX Hackathon

Başvurular 9 Eylül’de kapanıyor.

🥇 $6K
🥈 $3K
🥉 $1K

3 gün kaldı.

Yeni zincirler
Daha fazla builder
Daha fazla creator
Ve büyüyen bir agent economy.

Build the Era. 🟢

[asrin @asriintncy]

bizim kızlar milletler ligi ve avrupa şampiyonu oldu bu arada #FileninSultanları #BizimKızlar #kpss26 ÖSYM Ali Koç

[Hikmettuncay @karakalemhikmet]

This @axisrobotics update is bigger than it looks.

Challenger task distributions are moving to a fixed weekly schedule.

Every Monday at 16:00 SGT, 20 to 50 Challenger tasks will be distributed.

At first, this might look like a simple scheduling change.

I actually think it makes the whole system much more interesting.

Trainers can now plan their work around a predictable flow, while the research team can process contributions in structured batches.

And this comes right after Axis introduced a Challenger system focused on both volume and quality, with harder tasks earning 2 to 3x more points.

That combination matters.

We are moving from simply collecting more trajectories to building a more structured pipeline for collecting better trajectories.

The part I find most exciting is that every contribution can teach the robot something different.

Sometimes a few seconds of the right correction can be more valuable than an entire successful run.

More data does not always mean better data.

The right data, produced consistently and verified properly, is what moves Physical AI forward.

Monday is going to be interesting.

[asrin @asriintncy]

@connorcrd 0xF4E5EB6a3dc455aE82fb4539142746E21592e81B

[asrin @asriintncy]

Uğur karakullukçu programa kırmızı şort ile çıkmış Ersin Düzen boxer ile. Bu nasıl bir iş ahlakı ya.
Mehmet Şimşek Chemsdine Talbi Sercan Dikme Matteo Guendouzi Namaz #gemikazası

https://t.co/XlOjO1PocF

[Hikmettuncay @karakalemhikmet]

Guys, I took a deeper look at the latest @axisrobotics Axis Weekly and honestly, some of these updates surprised me. 👀

We’re here every day trying to complete tasks and control these robots, but sometimes we don’t realize how much is actually improving behind the scenes.

First, one important reminder:

Make sure your Axis account is connected to your Kaito account so your campaign activity is tracked properly.

If you’ve already completed the Kaito part, keep working through the tasks on Axis:

Axis page

Because as you complete tasks, you unlock badges.

And I wouldn’t look at these badges as just profile decorations.

To me, they’re more like a progress map for becoming a better contributor to the robot training dataset.

But the really interesting part is what’s happening underneath:

DAgger round 2 performance went from
68/160 → 78.3/160

That’s roughly a +10.3 improvement.

HG-DAgger candidates also gained 4–7 points over the baseline.

TaskGen can now break tasks into cleaner single-scene setups.

RoboCasa now has live interactions with drawers, cabinets and appliances.

And the sim-to-real hardware setup is now ready.

Now think about that for a second…

We’re sitting here trying to control a virtual robot and sometimes I still make it all the way to the final step just to lose the task. 😅

It happens to me almost every day.

But behind those attempts, there’s a much bigger system being built to generate training data that can help real robots perform better in the real world.

And this is where the referral update gets interesting.

Your score = Kaito mindshare × referral multiplier

So the multiplier actually matters.

The first milestones are now easier to reach:

5 qualified referred Trajectories → 1.05x
10 qualified referred Trajectories → 1.1x

But simply inviting someone isn’t enough.

They need to complete the task
pass Axis verification
and sign the transaction onchain.

So it’s not just about bringing people in.

It’s about bringing in contributors who actually complete the work.

That’s what makes Axis interesting to me.

We’re not just farming points.

We’re completing tasks
making mistakes
trying again
unlocking badges
and contributing to data that could eventually help train better robots.

That’s also why I keep waiting for the next task every day.

Even when I get so close and lose on the final step…

I genuinely want to see what happens on the next attempt. 🤖

Keep building.

And shoutout to @plpiaoliang for helping surface these details. 🙌

gAxis @axisrobotics
@axisroboticsTR

[asrin @asriintncy]

@bosversenecemre Aybala

[asrin @asriintncy]

Zafer Bayramımız kutlu olsun. 🇹🇷
başta Gazi Mustafa Kemal Atatürk olmak üzere tüm kahramanlarımızı saygı, minnet ve rahmetle

Aziz Başkan Saran Kaymakamımız Sn Ruhi Çenet Sahil Güvenlik #WeLoveYouSane Girne #ölüdeniz Türk Silahlı Kuvvetleri Günü Yusuf Cem Yılmaz Penaltı

[Hikmettuncay @karakalemhikmet]

I keep coming back to one thing about @axisrobotics

The real challenge in Physical AI isn’t collecting more robot data.

It’s making every new round of data more useful than the last.

Train a policy
Find where it fails
Collect targeted human corrections
Train again

That’s a much more interesting loop than simply stacking millions of trajectories.

And scale isn’t just about more data either.

If a model only works on one robot arm, that’s not real generalization.

Different morphologies
Different hands
Different bodies
Same skills

That’s where Physical AI gets interesting.

The future won’t be built by the biggest dataset.

It’ll be built by the best data flywheel.

[Axis Robotics @axisrobotics]

Axis has been building alongside @privy_io from Day 0.

As a pioneering Physical AI data engine in the Privy ecosystem, we’re proud to provide a friction-free onboarding experience for over 150,000 global contributors through social login.

Together, we’ve achieved:
✱ Over 3.7M Trajectories Collected
✱ 42,000+ Hours of Trajectory Data
✱ 4,000+ Active Tasks

Experience it yourself today: https://t.co/h1HaIkMiu7

[asrin @asriintncy]

Victor Osimhen, Anderson Talisca

[Hikmettuncay @karakalemhikmet]

Good morning gAxis 🦾🤖

The @axisrobotics x Dexmal AI partnership is more interesting to me than just another robotics announcement

What stands out is the role Axis is taking here

Dexmal is building general purpose physical intelligence while Axis provides the data infrastructure needed to push VLA and world models further

Large scale simulation
Real world data
Egocentric data
Human generated training signals

I think the real race in Physical AI won’t only be about building better robots

It will be about who can build the better data engine behind them

Robots generate data
Data improves models
Better models improve robots

That loop is where things get interesting

And I think this partnership shows why Axis is building something much bigger than a simulation platform 🦾

[Hikmettuncay @karakalemhikmet]

Axis is changing how contribution works

And one thing I find interesting is that every action here can become data

That data can eventually help train real robots

So when I complete a task, I’m not just farming points

I’m helping build Physical AI 🤖

I’m not the fastest contributor 😅

But every task teaches me something about controlling the robot better

A few things that helped me:

↑ ↓ ← → for movement
E / D for vertical movement
Q / W for rolling
A / S for tilting
Z / X for rotation
N to save a checkpoint
B to return to the last checkpoint
R to restart

The biggest lesson for me:

Slow down near the target

Watch both camera angles before moving

And if a grip fails, don’t waste time fighting it

Just go back and try again

What makes the current change even more interesting is the Kaito side

It’s not only about creating content anymore

Bringing real users who actually contribute matters too

Join through my Kaito link 👇
https://t.co/Wq1BbmjOda

Content → real users → more contributions → more training data

That’s a much more interesting loop for Physical AI

Slow progress is still progress when every task creates useful data